Dr. Koo raised an epidemiologically and clinically relevant issue regarding the definition of low HDL-C level. In defining metabolic syndrome, the cutoff level of low HDL-C is frequently higher for women because women have a higher HDL-C distribution compared to men. According to the definition of metabolic syndrome in the NCEP-ATP III guidelines, reduced HDL-C level is defined as <50 mg/dL in women and <40 mg/dL in men [3]. However, the NCEP-ATP III criteria for reduced HDL-C might not be appropriate for the Korean population. Compared to the Western population, East Asian populations have lower HDL-C level and show smaller sex differences for this measure. Moreover, HDL-C levels among Koreans are even lower than those of other East Asian populations and show only a small sex difference [5-8]. In our study, the sex difference in mean HDL-C (women minus men) was only 2.1 mg/dL in 1998, 3.6 mg/dL in 2001, and 5.0 mg/dL in 2005 [9]. Because Korean women have relatively low HDL-C levels, 58.5% of women had an HDL-C level lower than 50 mg/dL in the 2001 KNHANES [10]. This caused the unusually high prevalence of metabolic syndrome, in comparison to the low prevalence of obesity, among Korean women in the literature. We also lack direct data regarding the level of HDL-C that should be treated. It is unclear whether women with HDL-C level of 40 to 49 mg/dL but without other lipid abnormalities should be treated. For these reasons, some Asian studies have used a cutoff for reduced HDL-C of <40 mg/dL for both sexes [6,11].

Overall, I appreciate and agree with Dr. Koo's considerate comments. Great attention should be given to the working definitions when reporting and interpreting prevalence of metabolic syndrome and dyslipidemia. Further studies are also required to establish an appropriate definition of dyslipidemia in the Korean population.